© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 30164
    Prev
    Next
    INDEX
    Explanations

    numerical data or identifiers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_resid_post
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    23
    +0.16
    0.9%
    234
    +0.13
    0.7%
    496
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    23
    +0.16
    0.23
    496
    +0.13
    0.14
    464
    +0.12
    0.04
    Negative Logits
    weg
    -1.64
    ière
    -1.63
    haus
    -1.52
    aco
    -1.50
     Collaboration
    -1.50
    zilla
    -1.45
    lining
    -1.44
    /">
    -1.44
    tek
    -1.43
    idelines
    -1.43
    POSITIVE LOGITS
    ĻĤ
    5.14
    ľĵ
    4.83
    Ļª
    4.61
    ģ
    4.50
    Īĺ
    4.45
    »¿
    4.37
    Ĺ
    4.35
    ĥ
    4.33
    ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ
    4.23
    IJ
    4.16
    Activations Density 7.767%

    No Known Activations